Loading MCP server directory, please wait...
Detect prompt injection, jailbreak, and social-engineering attacks in LLM agents.
npx @getmcp/cli add io.github.dl-eigenart/agentshield-mcpRequires Node.js 18+. Community-contributed server. Review source before installing.
AGENTSHIELD_API_KEYYour AgentShield API key. Sign up at https://agentshield.pro/signup (free tier, no credit card).
Config file: ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"io.github.dl-eigenart/agentshield-mcp": {
"command": "npx",
"args": [
"-y",
"@eigenart/agentshield-mcp"
],
"env": {
"AGENTSHIELD_API_KEY": ""
}
}
}
}Reasonbook: decisions that survive the session, in a local record AI agents keep.
Start Claude Code and Codex agents the person can watch, wait for them and read their answers.
Turn any idea, URL, doc, or PDF into on-brand visuals: 100+ formats, native in 150+ languages
Turn footage into finished video: reframing, captions behind the subject, motion graphics, renders.